I really enjoyed reading The Wisdom of Crowds, by James Surowiecki. The premise of the book is that, under many circumstances, large groups of people make better decisions than a few experts. He argues that the group is even smarter than the smartest individual in the group, and provides many good examples. One of the first examples is the counting of jelly beans in a jar; the average of all of the guesses will be closer to the truth than the closest answer of any individual. Of course, everyone can think of examples where crowds (or committees) have made very poor decisions; the book describes the circumstances in which groups of people are smarter than individuals.

To make good decisions, a group needs diversity. A group can know as much as all of the individuals in the group put together. Different perspectives make it most likely that the best solution will be found, because diversity adds new information to the group. (This topic applies directly to hiring practices; by hiring people that are different than the people you already have, you're making the group even smarter, even if the new people aren't as skilled.)

Another requirement is independence. The best decisions are made by groups of people that aren't simply "following the leader." It is important that some members of the group "buck the trend," even if their opinions are wrong, not only because the "errors" can average out in the end, but because reserved people with good opinions are more likely to speak up if they see others doing so.

Decentralization feeds both diversity and independence; when people are spread out, they are more likely to come up with different solutions. The final key component is thus aggregation; you need a way to collect all of the solutions of all of the members of the group to generate the final answer.